Amplify GenAI Models

Amplify offers a variety of models; Carleton has chosen to select a few from this list to make choice for users easier. Please note that not all models have the same capacities and that these models are not always the most recent models available for premium pay. This page gives an overview of what each model provides and also a quick overview over what features may work better in what model.

Only the models currently active in Amplify will have a brief description below. However, more models are possible through the Amplify platform, and are listed in a table found at the end of this page.

Generative AI Models (March 2025)

Text is borrowed from Amplify’s documentation.

• GPT-4o is good for complex tasks requiring advanced understanding. As GPT-4o is the model available to paid ChatGPT Plus users, and to a more limited extent free users, it offers further advanced intelligence over its predecessors. GPT-4 can also carry out complex mathematical operations, code assistance, analyze intricate documents and datasets, demonstrates critical thinking, and in-depth context understanding. This model is trained on information available through April 2023.
• Claude 3.5 Haiku is useful for ultra-low-latency applications that still need advanced reasoning, coding, and multilingual support. Claude 3.5 Haiku is Anthropic’s fastest model, matching or exceeding Claude 3 Opus accuracy on many benchmarks while keeping costs minimal. Handles text while offering 200 k-token context, and is ideal for real-time chat, code autocompletion, and content moderation. Trained on information available through July 2024.
• Claude 4 Sonnet is considered Consider for tasks requiring a balance between speed and complex reasoning. Anthropic’s Claude 4 Sonnet has “hybrid reasoning,” allowing users to choose between near-instant responses and extended, step-by-step analysis. This flexibility enhances performance in coding, mathematics, and intricate problem-solving. The model also features a “scratchpad” that displays its reasoning process, aiding in understanding and refining prompts.

Below you will find the remaining models available to Amplify. For more information about access to these models, please contact Don Vosburg (dvosburg) in Academic Technology.